How can I help you solve your challenges related to deployment/operations/devops?
I’m the author of the Passenger app server. We’re on a mission to make Unix server tools easier to use, more convenient, more robust and generally more awesome. So we are thinking about what we should work on next.
I would like to ask you all - what kind of problems do you encounter when using Ruby/Rails, or when dealing with deployment/operations/devops in general, that need solving? Or what do you experience as cumbersome or suboptimal and think needs improvement? How important do you deem those issues to be, and how satisfied you are with existing solutions/mitigations/workarounds for them?
Please let me know either by replying here, by emailing me (email@example.com) or sending me a tweet (@honglilai). Even if you yourself don’t have any challenges or ideas, please forward this message to your friends. I would like to hear their feedback as well!
To me the most painful part is the management of long running processes (sidekiq workers, etc.) They feel so badly integrated you always have to add forman, integrate it with capistrano, generate config files for your OS’s preferred process management daemon, which of course can change, the templates doesn’t support the features from one tool to another, etc.. I would love to have, say, passenger auto spawn (and scale) my sidekiq processes like there were web workers, restart them when they crash, etc… It could also monitor simple unique background processes to eliminate the need for foreman and OS dependent tools.
Post a comment